WEBVTT 00:00:00.000 --> 00:00:05.200 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds. 00:00:05.340 --> 00:00:07.800 This is episode 478. 00:00:08.200 --> 00:00:09.200 I am Michael Kennedy. 00:00:09.560 --> 00:00:10.240 And I'm Brian Okken. 00:00:10.500 --> 00:00:18.540 This episode is recorded on May 4th, 2026, brought to you by us and all the things we're doing. 00:00:18.700 --> 00:00:20.980 I'm sure we'll mention something during the show somehow. 00:00:21.260 --> 00:00:23.260 I've got a few cool updates to discuss. 00:00:23.580 --> 00:00:28.020 If you want to connect with us on social media, maybe submit a topic that you think we should pay attention to. 00:00:28.020 --> 00:00:30.360 Check us out on all the social medias. 00:00:30.480 --> 00:00:34.000 You can find us on the show page, episode page. 00:00:34.100 --> 00:00:36.500 Pretty much everyone has a link to all the things there. 00:00:36.820 --> 00:00:37.680 Sign up for the newsletter. 00:00:38.220 --> 00:00:42.740 It adds way more than just show notes or, hey, we published an episode. 00:00:42.920 --> 00:00:45.020 It's got a bunch of cool extra information that we put in there. 00:00:45.080 --> 00:00:46.180 So check that out. 00:00:46.540 --> 00:00:50.920 And with that, Brian, I think we should get started. 00:00:51.140 --> 00:00:51.920 What's your first topic? 00:00:52.220 --> 00:00:54.240 I'm going to talk about profiling explorer. 00:00:54.240 --> 00:01:03.160 So this is a profiling explorer is a tool from Adam Johnson of lots of things Django and Git related fame. 00:01:03.660 --> 00:01:07.580 So Adam, so how did I see this? 00:01:07.760 --> 00:01:08.020 I don't know. 00:01:08.140 --> 00:01:08.640 Oh, yeah. 00:01:08.700 --> 00:01:14.300 He had a blog article that he released last month that somehow I just found out. 00:01:14.400 --> 00:01:15.240 I read today. 00:01:15.240 --> 00:01:18.780 Introducing profiling explorer. 00:01:19.660 --> 00:01:20.900 Actually, I picked it up yesterday. 00:01:21.040 --> 00:01:24.120 But anyway, so this is kind of fun. 00:01:24.280 --> 00:01:26.240 This is a says I've made it another package. 00:01:26.580 --> 00:01:26.700 Wait. 00:01:27.040 --> 00:01:27.440 Anyway. 00:01:27.440 --> 00:01:30.340 So what do we have here? 00:01:30.400 --> 00:01:32.800 We have a way to profile. 00:01:33.060 --> 00:01:36.660 We have lots of ways to profile Python code. 00:01:36.660 --> 00:01:46.600 But this takes the output of the profiler, the C profiler, and pops it into a table that you can interact with. 00:01:46.900 --> 00:01:50.940 So this lovely table that also has dark mode. 00:01:51.000 --> 00:01:53.320 He shows it in light mode, but it also has dark mode. 00:01:53.320 --> 00:01:58.900 And it's got all the normal stuff you'd think of for profiling. 00:01:59.060 --> 00:02:09.580 You've got the cumulative time, the percentage of your time in a certain place, and the number of calls that go to this place or this thing. 00:02:09.940 --> 00:02:12.480 So but you can filter on the filters kind of cool. 00:02:12.600 --> 00:02:25.160 So you can click on the calls or the internal time or cumulative time column headers to sort by that, which that's where I usually start is certain by the percentage of the time spent somewhere. 00:02:26.060 --> 00:02:29.500 Anyway, then you can also do there's a search box. 00:02:29.500 --> 00:02:41.540 So you can filter on like if you like even if you grab all of the time, but you're really looking at a subcomponent that you're working on, you can just look at that and see where it's small. 00:02:41.540 --> 00:02:52.840 So like that's often the time where I'm often grabbing, looking at the profiling documentation is how do I filter down to just the thing I want to care about? 00:02:53.180 --> 00:02:55.300 And sometimes it's easier just to grab it all. 00:02:55.440 --> 00:02:59.520 And then and then just using this, you could filter in exactly what you want. 00:02:59.800 --> 00:03:01.540 So yeah. 00:03:01.800 --> 00:03:04.280 And there's a hover features and caller and colleagues. 00:03:04.640 --> 00:03:08.520 And I guess look through the article to figure out what what this is all about. 00:03:08.560 --> 00:03:09.600 But it looks really slick. 00:03:09.600 --> 00:03:12.600 I haven't tried it yet, but I'm going to probably this week. 00:03:12.820 --> 00:03:13.560 So it looks fun. 00:03:13.800 --> 00:03:16.580 And profiling is certainly finding a needle in a haystack. 00:03:16.700 --> 00:03:19.600 So all the UI, all the search you can get is very good. 00:03:19.860 --> 00:03:20.060 Yeah. 00:03:20.240 --> 00:03:26.160 And there's like profiling often has like there's some people that like the flame graphs. 00:03:26.340 --> 00:03:29.720 I've never gotten the flame graphs to work in my brain. 00:03:29.820 --> 00:03:30.820 I just don't get them. 00:03:31.080 --> 00:03:34.600 So so something like a table works for me. 00:03:34.740 --> 00:03:35.280 I like this. 00:03:35.720 --> 00:03:35.900 Yeah. 00:03:35.900 --> 00:03:39.340 I really like the call call hierarchy type of thing. 00:03:39.340 --> 00:03:44.960 It's almost the call call stack, but it's split if it takes different branches that you see in like PyCharm and stuff. 00:03:45.180 --> 00:03:45.380 Yeah. 00:03:45.440 --> 00:03:45.940 That works for me. 00:03:45.980 --> 00:03:47.560 But still, those things can be too noisy. 00:03:47.780 --> 00:03:48.740 So very cool. 00:03:48.960 --> 00:03:49.120 Yeah. 00:03:49.320 --> 00:03:49.560 Nice. 00:03:49.820 --> 00:03:50.100 Nice. 00:03:50.100 --> 00:03:55.840 Now let's bump over and talk about reverting the incremental GC in Python. 00:03:55.840 --> 00:03:56.720 Like what? 00:03:56.940 --> 00:04:04.660 So this is a, I don't know who just, I think just the core developers and steering council as a group decided on this. 00:04:04.740 --> 00:04:07.000 But the post is by Hugo Van Commande. 00:04:07.840 --> 00:04:14.920 And it's just entitled reverting the incremental GC in Python 3.14 and 3.15, which is pretty wild. 00:04:14.920 --> 00:04:17.440 And there's, this is on discuss.python.org. 00:04:17.520 --> 00:04:19.980 And so there's a bunch of back and forth here. 00:04:20.240 --> 00:04:21.820 So let me give you some of the highlights. 00:04:21.960 --> 00:04:22.100 Okay. 00:04:22.260 --> 00:04:26.460 So Python 3.14 shipped with a new incremental garbage collector. 00:04:26.680 --> 00:04:31.620 So I believe the way it worked previously is when a garbage collection happened. 00:04:31.760 --> 00:04:34.000 And maybe I should point out why garbage collections happen. 00:04:34.000 --> 00:04:39.460 So Python is a little weird in that it has two types of memory management. 00:04:39.880 --> 00:04:40.700 Both of them automatic. 00:04:40.920 --> 00:04:42.020 You don't worry about them. 00:04:42.260 --> 00:04:44.400 But they're two ways of managing memory. 00:04:44.520 --> 00:04:46.160 And they happen in different scenarios. 00:04:46.360 --> 00:04:46.460 Okay. 00:04:46.520 --> 00:04:51.220 So the primary, by far, way that memory management happens is by reference counting. 00:04:51.520 --> 00:04:52.140 You get a variable. 00:04:52.540 --> 00:04:53.260 It has one reference. 00:04:53.620 --> 00:04:54.480 You pass it to a function. 00:04:54.640 --> 00:04:55.560 Now it has two references. 00:04:55.800 --> 00:04:58.000 That function puts it into a list, but then returns. 00:04:58.220 --> 00:05:00.200 It goes up to three and then back to two. 00:05:00.520 --> 00:05:01.560 You throw away the list. 00:05:01.900 --> 00:05:03.040 And then the variable goes away. 00:05:03.040 --> 00:05:03.080 Okay. 00:05:03.320 --> 00:05:04.460 Reference count hits zero. 00:05:04.640 --> 00:05:04.880 Delete. 00:05:05.040 --> 00:05:05.180 Right? 00:05:05.420 --> 00:05:06.940 Reference counting works great. 00:05:07.060 --> 00:05:07.940 It's super fast. 00:05:08.040 --> 00:05:08.600 It's immediate. 00:05:08.860 --> 00:05:12.700 Except it does not work great when you have a cycle. 00:05:13.060 --> 00:05:13.200 Right? 00:05:13.340 --> 00:05:17.780 I've got something, two objects that point to each other for some reason. 00:05:18.080 --> 00:05:20.200 All of a sudden, their reference counts never hit zero. 00:05:20.560 --> 00:05:21.760 Memory goes through the roof. 00:05:21.980 --> 00:05:22.380 You're toast. 00:05:22.700 --> 00:05:35.960 So what we're talking about is the thing, the garbage collector that doesn't use reference counting, but actually traverses all of the references and then looks for those cycles that are abandoned, but themselves never went to zero and collects it. 00:05:36.000 --> 00:05:36.100 Okay. 00:05:36.180 --> 00:05:42.380 So there was a new incremental garbage collector added that the way it worked now is you would have different generations. 00:05:42.540 --> 00:05:43.880 Those generations would run. 00:05:44.460 --> 00:05:47.260 When it was time for a collection, the entire app stopped. 00:05:47.640 --> 00:05:49.340 Garbage collection was checked. 00:05:49.580 --> 00:05:50.920 These cycles were checked. 00:05:51.080 --> 00:05:52.180 And then the app carried on. 00:05:52.180 --> 00:05:55.280 So the incremental one tries to improve latency. 00:05:55.720 --> 00:05:58.520 Say like, look, we don't have to stop the program entirely. 00:05:58.520 --> 00:06:03.880 We're going to do a little bit of garbage collection, a little bit more running, a little bit more garbage collection and so on. 00:06:04.080 --> 00:06:06.900 Of course, the picture of garbage could change over that period. 00:06:06.900 --> 00:06:07.120 Right? 00:06:07.320 --> 00:06:08.580 Anyway, that was added. 00:06:08.580 --> 00:06:13.540 But it turns out production reports started to come in for severe memory pressure. 00:06:14.100 --> 00:06:19.920 Neil measured up to five times peak memory on pathological cyclic workloads. 00:06:20.080 --> 00:06:23.880 And cyclic is the important part because that's actually where GC actually does anything. 00:06:24.300 --> 00:06:26.320 It actually makes, you know, actually does the cleanup. 00:06:26.640 --> 00:06:29.720 And that pushed the core team and steering council to revert it. 00:06:30.060 --> 00:06:33.220 It's in 314, but it's going to be taken out of 314. 00:06:33.460 --> 00:06:37.340 And it's going to not actually make it to 350, which is pretty extreme, I think. 00:06:37.340 --> 00:06:37.580 Right? 00:06:37.820 --> 00:06:41.920 It takes a lot to get something into Python, but once it's in, it's rarely taken out. 00:06:42.160 --> 00:06:42.260 All right. 00:06:42.260 --> 00:06:45.200 So this is actually the second time the incremental GC has been pulled back. 00:06:45.360 --> 00:06:48.360 It was reverted right before 313 first came out. 00:06:48.580 --> 00:06:49.800 As they were like, oh, is that quite ready? 00:06:49.980 --> 00:06:53.540 And then it got into 314 without going through the PEP process. 00:06:53.640 --> 00:06:55.180 So this is a bit of the tricky stuff. 00:06:55.400 --> 00:06:57.700 However, the benefit was real. 00:06:57.780 --> 00:06:59.560 So like they were trying to reduce latency. 00:06:59.760 --> 00:07:07.780 So with the incremental one, the max GC pause times for one example was one millisecond versus 26 milliseconds with the old one. 00:07:07.980 --> 00:07:08.700 So that's great. 00:07:08.760 --> 00:07:14.220 Except for it was just blowing up the memory usage of things that cared more about memory and less about latency. 00:07:14.640 --> 00:07:20.380 And a lot of places we run Python care more about memory than they care about very, very small pauses. 00:07:20.800 --> 00:07:25.320 You know, think like APIs and web apps and anything that runs on a server or in the cloud. 00:07:25.440 --> 00:07:28.200 Memory is probably the most expensive resource. 00:07:28.380 --> 00:07:28.580 Right? 00:07:28.580 --> 00:07:29.620 So that's not great. 00:07:29.760 --> 00:07:33.700 So Hugo decided we're going to ship 314.5 early with the revert. 00:07:33.980 --> 00:07:43.020 And Gregory Smith floated the idea of a 314.5 RC1, a release candidate for a shipping version of Python, which would be interesting. 00:07:43.300 --> 00:07:43.460 Okay. 00:07:43.640 --> 00:07:50.960 And then Tim Peters jumped into thread, doing a bunch of live forensics on Windows, running a toy DQ program that should cap at one gigabyte. 00:07:51.140 --> 00:07:55.860 And it used up to 15.6 gigs on a 16 gig machine. 00:07:56.160 --> 00:07:57.540 So that's not great. 00:07:57.540 --> 00:07:58.920 And this is really bad. 00:07:59.100 --> 00:08:04.020 He discovered that Gen Zero collections effectively never fired under this new collector. 00:08:04.900 --> 00:08:08.080 That's the one that gets most of them and it's the cheapest to run. 00:08:08.180 --> 00:08:09.280 So that's a big problem. 00:08:09.540 --> 00:08:18.840 Anyway, this might sound bad, but Tim points out that CPython has had a chronic shortage of real world GC benchmarks. 00:08:19.240 --> 00:08:29.500 High performance, the test suite that they run Python on for performance considerations, says it has basically no interesting cyclic workloads and users almost never share real data. 00:08:29.680 --> 00:08:31.080 So core devs keep flying blind. 00:08:31.480 --> 00:08:34.400 And circling back to your previous topic, right? 00:08:34.620 --> 00:08:46.240 Is Adam Johnson published a blog post along the thread of a real memory leak in Django migration systems caused by the incremental GC with where you have to call GC collect manually to fix it. 00:08:46.240 --> 00:08:46.600 All right. 00:08:46.680 --> 00:08:49.300 So it may or may not come back for 3.16. 00:08:49.300 --> 00:08:52.680 If it does, it was going to be restarted through the PEP process. 00:08:53.260 --> 00:08:54.660 So a lot going on here. 00:08:54.880 --> 00:08:55.580 That's crazy, right? 00:08:55.880 --> 00:08:56.420 Well, yeah. 00:08:56.480 --> 00:09:10.900 But I also hope that, I mean, Tim's comment is like, we kind of flew past it, but it's really, it's really important that the cyclic workload, having some of those in a benchmark is important. 00:09:11.320 --> 00:09:18.220 But if you can come up with a pathological way where it doesn't work as well, it shouldn't crash. 00:09:18.380 --> 00:09:27.640 It shouldn't be like, even an incremental should eventually go, you know what, we need to shut down and clean things up if necessary, I think. 00:09:27.640 --> 00:09:41.340 but, but we also need to have a benchmark that's more realistic and it's hard to tell what realistic is because we're doing, Python's being used for everything from, you know, from, from particle physics to web pages. 00:09:41.460 --> 00:09:43.040 so who knows? 00:09:43.200 --> 00:09:44.840 I have a suggestion for them. 00:09:44.900 --> 00:09:45.640 I agree with you. 00:09:45.940 --> 00:09:46.880 I have a suggestion. 00:09:47.080 --> 00:09:50.460 One of the areas this shows up very often is in ORMs. 00:09:50.860 --> 00:09:51.360 Oh yeah. 00:09:51.680 --> 00:09:51.900 Right. 00:09:51.900 --> 00:10:00.180 So imagine I've got a user and, maybe a course and then a user activity, right? 00:10:00.240 --> 00:10:13.380 So if I go run a query, you can send a database for all those things that are in, maybe the user activity has a user ID, but it also has a like reverse lazy loaded back to the user itself and back to the course itself. 00:10:13.380 --> 00:10:15.560 But the user's holding onto their activity. 00:10:15.560 --> 00:10:26.760 Cause they've got like a list of activities, which was also a lazy loaded and you like tell SQLAlchemy or something like we're going to do eager loading and all these things does creates tons of cycles just by doing database queries. 00:10:26.860 --> 00:10:32.840 Like I've seen that kind of stuff where the GC makes a difference on my web apps that actually do a lot of database things. 00:10:32.840 --> 00:10:42.940 So maybe they could come up with like a real interesting web app database query sort of example, and just use put SQL, a static SQLite database that's in there and integrate that. 00:10:42.940 --> 00:10:50.780 I mean, I'm sure there's a hesitancy to say, well, the way you run PI performance is you're going to fire up this database and this Redis queue. 00:10:50.920 --> 00:10:55.600 And then you're going to like, you know, like the infrastructure doesn't make any sense, but you could do it with SQLite. 00:10:55.680 --> 00:10:59.480 And I think you could, you could come up with some interesting, useful examples that have tons of data. 00:10:59.480 --> 00:11:01.000 Cause just fill the database, you know? 00:11:01.180 --> 00:11:01.380 Yeah. 00:11:01.580 --> 00:11:01.860 Anyway. 00:11:02.240 --> 00:11:02.640 Interesting. 00:11:02.760 --> 00:11:04.100 I think it's worth considering. 00:11:04.280 --> 00:11:05.600 It is worth considering. 00:11:06.040 --> 00:11:08.180 Oh, I got to hop over to my next topic. 00:11:08.480 --> 00:11:11.360 Nick's this, speaking of discussion. 00:11:11.360 --> 00:11:15.740 So this is just, I guess, I guess I'm just like passing along a discussion. 00:11:15.740 --> 00:11:17.780 I was amused to, to read about. 00:11:18.100 --> 00:11:25.580 so this is, few, I, it changed today, yesterday that the topic changed. 00:11:25.580 --> 00:11:33.700 So I was surprised to find out that Microsoft, and VS Code enabled AI coauthor by default. 00:11:34.300 --> 00:11:37.700 So, and I'm, I'm like, what really? 00:11:37.700 --> 00:11:52.060 So I read a little bit more about this and really it's, so you can, you can do get commits and stuff to that have, whether or not you've, you know, whether or not AI contributed to the code. 00:11:52.120 --> 00:12:00.720 And instead you used to be able to just like say, it used to be only of like get or your AI is actually committing. 00:12:00.720 --> 00:12:07.120 but, but, you know, copilot wants to get credit if they're doing a lot of work, I guess. 00:12:07.140 --> 00:12:07.700 I don't know. 00:12:07.800 --> 00:12:11.920 But, so, there's a flag that you can set. 00:12:11.920 --> 00:12:16.740 That's the get dot add, add AI coauthor configuration. 00:12:16.980 --> 00:12:20.900 And there, it was started with off and they defaulted. 00:12:21.160 --> 00:12:23.160 They won't change it to default of all. 00:12:23.160 --> 00:12:36.340 and, it isn't all the time, but if there's something in here that tries to detect whether or not you used any, any of the code was, contributed by an AI workflow or anything. 00:12:36.340 --> 00:12:49.300 And I'm not sure about all the details, but people freaked out basically because like, I was like, even if I, even if I run a, like a copilot to try to figure out something, I may 00:12:49.300 --> 00:12:56.720 have rejected everything or, or something and, and, or maybe my code looks like something that the, I don't know, how does it detect it? 00:12:56.740 --> 00:12:57.240 I'm not sure. 00:12:57.640 --> 00:13:00.180 and, and I don't think that's right. 00:13:00.260 --> 00:13:02.680 And basically I just don't think it's right. 00:13:02.680 --> 00:13:04.540 Well, I was about to talk about that. 00:13:04.660 --> 00:13:13.580 And then I noticed that I was reading the comments this morning and found out that there was a follow-up and the follow-up was just yesterday, changed the default back to off. 00:13:13.580 --> 00:13:18.700 so, and then, and then it devolved. 00:13:18.840 --> 00:13:25.860 They closed the comments, even in one day, they closed the comments because basically the comments turned into, why are you even using VS Code? 00:13:25.940 --> 00:13:32.200 You should use VI or, you know, Emacs or, or what are the other, some of the others and stuff. 00:13:32.200 --> 00:13:39.880 So anyway, I use, I use, I use a lot of editors, not a lot, a couple, but VS Code is one of them. 00:13:39.880 --> 00:13:44.220 and I was the, I think the main takeaway from this is I don't trust them. 00:13:44.320 --> 00:13:49.480 I'm going to go in my settings and, go ahead and whatever this, no, I can't find it. 00:13:49.580 --> 00:13:52.420 The setting was, was like, whatever. 00:13:52.580 --> 00:13:53.300 Oh, here it is. 00:13:53.380 --> 00:13:55.240 The, the ad AI coauthor. 00:13:55.420 --> 00:14:03.260 I'm going to just go ahead and set that to off in, in my, in my settings file, because I don't trust them to turn it back on later. 00:14:03.900 --> 00:14:06.020 anyway, I think this is crazy. 00:14:06.020 --> 00:14:09.780 First of all, let me defend the people who say VS Code. 00:14:09.880 --> 00:14:12.840 Where, where folks are saying you should use them or Emacs. 00:14:13.260 --> 00:14:19.920 I'm not, I feel like a lot of those editors encourage bad, bad design practices, right? 00:14:20.120 --> 00:14:23.500 Whereas proper IDE like things are a little bit better. 00:14:23.660 --> 00:14:24.940 So I think that that's fair. 00:14:25.040 --> 00:14:33.240 I don't think they should take that personally, but this AI coauthor stuff is just a, so that was kind of put the defense up before we started hacking on them. 00:14:33.440 --> 00:14:36.340 So I think I, this is the first time hearing this, Brian, this is, this is garbage. 00:14:36.340 --> 00:14:39.300 Now I haven't noticed this because I don't use copilot. 00:14:39.300 --> 00:14:50.100 Cause I have more respect for my work time and just use cloud code, in some kind of VS environment with the extension to be clear. 00:14:50.300 --> 00:14:53.980 And I also hate it if cloud does it, but every now and then cloud does it. 00:14:53.980 --> 00:14:55.300 And it drives me crazy. 00:14:55.540 --> 00:15:01.440 If I ask it for a commit message, it'll sometimes give me a bunch of stuff and then coauthored by cloud code. 00:15:01.440 --> 00:15:05.700 And it'll like at mention their GitHub username or whatever. 00:15:05.700 --> 00:15:07.580 And like delete, delete, delete, delete, delete. 00:15:07.580 --> 00:15:09.780 So I've never seen this. 00:15:09.840 --> 00:15:11.320 So where does it show up? 00:15:11.360 --> 00:15:14.320 Does it, is it like behind the scenes or does it show up? 00:15:14.320 --> 00:15:18.620 The most common place I've seen it is in pull requests, but I've also seen it in commit messages. 00:15:18.800 --> 00:15:20.020 So let's just say it's a pull request. 00:15:20.020 --> 00:15:20.300 Okay. 00:15:20.560 --> 00:15:23.300 So you might go, I've not, is it in the message though? 00:15:23.480 --> 00:15:23.820 Yes. 00:15:24.120 --> 00:15:24.320 Yes. 00:15:24.320 --> 00:15:33.580 So I've, I've not used it coauthor, like I copilot, like I just said, but I've gone to cloud and I've said, Hey, you know, there's a bunch of stuff going on here. 00:15:33.580 --> 00:15:35.980 And I kind of forgot all the details that changed. 00:15:35.980 --> 00:15:49.720 Like, wouldn't it be a little more thorough if I said, Hey, Claude, look at all the pending changes on this branch since I've branched it and give me a nice summary and overview of actually what's changed instead of the stuff that I just intended to change. 00:15:49.720 --> 00:15:57.760 Maybe something else, you know, like maybe I ran the linter and it changed a bunch of stuff, but when I really only changed one line, but I would like to talk about, you know, whatever, something like that. 00:15:57.800 --> 00:16:12.380 It will say, here's the title, here's the description that you put, and it'll say all of it's describing and the bottom, it'll have a last single sentence paragraph that says coauthored by Claude code at anthropic or at Claude or something like that. 00:16:12.480 --> 00:16:13.200 That's lame. 00:16:13.460 --> 00:16:14.340 I'm like, screw this. 00:16:14.340 --> 00:16:16.160 So then I'll delete that, which is fine. 00:16:16.160 --> 00:16:20.640 But you can also just say, Claude, create a pull request on GitHub for me. 00:16:20.640 --> 00:16:22.000 And it'll do that beautifully. 00:16:22.180 --> 00:16:24.920 But you know what shows up at the bottom of that pull request description? 00:16:25.160 --> 00:16:26.220 Coauthored by Claude. 00:16:26.560 --> 00:16:34.720 So Claude doesn't do as often, but it sounds like, you know, Microsoft has gone absolutely bonkers over AI-ing everything. 00:16:35.020 --> 00:16:36.880 And I think that is a huge mistake. 00:16:36.880 --> 00:16:48.500 I think they have 40 different products named CoPilot that are actually different things like in Office 365 CoPilot, Windows CoPilot, CoPilot on GitHub, CoPilot and VS Code. 00:16:48.620 --> 00:16:48.880 You know what I mean? 00:16:48.920 --> 00:16:59.560 There's just like, there was a mandate that everybody, you know, a few years ago, 2023 or something like that, like every team that builds something at Microsoft needs to have some form of AI in it. 00:16:59.780 --> 00:17:01.980 And those were all independent projects and initiatives. 00:17:02.240 --> 00:17:04.000 And the adoption isn't that great. 00:17:04.000 --> 00:17:14.420 So this to me feels like a poor attempt, a bad attempt at kind of trying to growth hacking CoPilot back into the conversation, right? 00:17:14.500 --> 00:17:14.820 Yeah, maybe. 00:17:15.280 --> 00:17:15.460 Yeah. 00:17:15.520 --> 00:17:19.000 Think of like early days of Hotmail, which is completely forgotten. 00:17:19.140 --> 00:17:27.560 But one of it's really the things that was kind of a secret to a success, this is common now, but it wasn't then, is every email you sent was sent by Hotmail. 00:17:27.720 --> 00:17:30.320 Get your own one gigabyte free web email. 00:17:30.320 --> 00:17:33.540 And every single email had a link back to it. 00:17:33.760 --> 00:17:35.120 And people go, oh, you can have web email. 00:17:35.220 --> 00:17:35.600 That's cool. 00:17:35.740 --> 00:17:38.140 And like, like that was, and I feel like it's that. 00:17:38.160 --> 00:17:39.440 Social media stuff too. 00:17:39.560 --> 00:17:44.360 People would be like, you know, this post written on Mona on iPhone or something like that. 00:17:44.380 --> 00:17:45.120 Yes, yes, exactly. 00:17:45.560 --> 00:17:45.840 Yeah. 00:17:45.840 --> 00:17:46.500 There's a lot of that. 00:17:46.580 --> 00:17:49.380 So I think it's, it's got a strong ick factor. 00:17:49.380 --> 00:17:54.680 I just checked my default VS Code, which I don't know whatever it was, but it had this on. 00:17:54.780 --> 00:17:57.840 And so I just turned it off and I'm going to go do a thorough cleansing. 00:17:58.240 --> 00:18:02.560 But I think it's a tree falls in the forest, but no one, there's a here sort of example for me. 00:18:02.580 --> 00:18:03.840 Cause like, I don't use CoPilot. 00:18:04.040 --> 00:18:11.080 So if I did, it would put the marking on the co, the co, attribution on it, but I'm not going to use it. 00:18:11.120 --> 00:18:11.820 So it's not a problem. 00:18:12.180 --> 00:18:19.140 I actually don't get the CoPilot hate because I I'm using it, but I've been using it 00:18:19.140 --> 00:18:29.620 from the start and it's just like mostly code completion and, and a, you can open a window to, to open AI chat if you want. 00:18:29.620 --> 00:18:34.300 But, the, the, whatever settings I have are not intrusive. 00:18:34.360 --> 00:18:38.560 They were, they were intrusive for a while and I did something and they're not that bad anymore. 00:18:38.560 --> 00:18:42.380 And the comments, the suggested comments don't get in the way. 00:18:42.380 --> 00:18:45.000 And, and I'm, I'm, I appreciate it. 00:18:45.000 --> 00:18:48.740 And the, and I do use VS Code, but I also use Vim mode. 00:18:49.080 --> 00:18:56.400 so I, it's basically a, a big wrapper around Vim as far as like I'm using it. 00:18:56.480 --> 00:19:05.220 I do like the other thing I really like about it is, the integration with get so that I can, I can find out all the stuff I've changed and use do diffs easy and stuff. 00:19:05.220 --> 00:19:06.840 So I do like it. 00:19:06.840 --> 00:19:14.500 And also there's a lot of people using cursor and you can't hate on VS Code and like cursor because they're the same thing. 00:19:14.660 --> 00:19:18.020 so they're not exactly the same thing, but it's a fork. 00:19:18.020 --> 00:19:22.980 So anyway, a hundred percent, I turn off the autocomplete for all those tools. 00:19:23.080 --> 00:19:30.340 I don't like them, but so my, my view into this world is just the agents, you know, it's been, let's, let's move on to the next topic, Brian. 00:19:30.400 --> 00:19:32.520 It's been, Ooh, you wouldn't know it. 00:19:32.520 --> 00:19:34.440 We had literally a heat advisory. 00:19:34.740 --> 00:19:36.480 It was almost 90 degrees yesterday. 00:19:36.700 --> 00:19:37.180 It was terrible. 00:19:37.180 --> 00:19:39.000 Boy, was it freezing in Django land. 00:19:39.100 --> 00:19:39.700 Let me tell you. 00:19:40.740 --> 00:19:43.120 So I want to talk about something called Django freeze. 00:19:43.240 --> 00:19:44.760 And I think this is a super cool project. 00:19:44.760 --> 00:19:47.620 It doesn't have very much attention. 00:19:48.020 --> 00:19:51.080 It's only got 117 stars and 16 forks. 00:19:51.360 --> 00:19:56.560 It's the kind of project that's really low risk, like really low risk, but has a ton of benefits. 00:19:56.940 --> 00:20:01.920 So the idea with Django freeze is I've written a Django app and I'm thinking about how do I deploy it? 00:20:01.920 --> 00:20:05.120 Maybe it's driving its content from the database. 00:20:05.420 --> 00:20:11.440 Like it's got a, it's like an e-commerce store and it's got, here's all your categories and here's the products you're selling and so on. 00:20:11.580 --> 00:20:15.200 Those come out of the database, but maybe they change infrequently or whatever. 00:20:15.540 --> 00:20:22.340 So you could use Django freeze to just convert a static HTML version of your site and deploy that somewhere. 00:20:22.460 --> 00:20:28.560 Have all the categories, have all the products, have all the little backlinks and all that kind of things that you might have, whatever your app does. 00:20:28.560 --> 00:20:34.120 But the deployment and operation, the DevOps side is static files, which is incredible. 00:20:34.540 --> 00:20:34.840 Oh, cool. 00:20:35.060 --> 00:20:36.100 It's good for operational. 00:20:36.300 --> 00:20:39.840 Like your static files can't go down unless just your server doesn't work anymore. 00:20:40.080 --> 00:20:42.940 You know, like there's very few things that can go wrong. 00:20:43.300 --> 00:20:52.540 You know, it's not like you're going to, well, I ran out of memory or there is, I don't know, like some kind of issue between the caching and like just all the issues you have running real websites. 00:20:52.680 --> 00:20:54.400 Like with static sites are so delightful. 00:20:54.740 --> 00:20:59.020 Put them up, log as the server's online, things are good, you know, and you can host them for free in many places. 00:20:59.160 --> 00:21:04.580 So if you've got a Django site and you think really it kind of could be static site, but I like to work in Django. 00:21:04.680 --> 00:21:07.800 I don't want to work in Hugo or something like that where it's foreign or weird. 00:21:07.900 --> 00:21:09.840 Hugo is a really weird, even though it is good. 00:21:09.840 --> 00:21:16.020 You could just write in one, write in Python, write in your favorite web framework, but then turn it into a static site. 00:21:16.060 --> 00:21:17.400 And I really think this is a cool idea. 00:21:17.660 --> 00:21:22.140 It's, you know, sort of a peer of Flask Freeze, I think is the name. 00:21:22.380 --> 00:21:23.220 Yeah, I think it's Flask Freeze. 00:21:23.520 --> 00:21:23.840 What do you think? 00:21:23.920 --> 00:21:24.220 Cool, right? 00:21:24.560 --> 00:21:24.920 Yeah. 00:21:25.020 --> 00:21:34.060 And one of the things I think one of the places where I'd probably, that I'm actually thinking of using this, I'll try it out, is like a SAS, for instance. 00:21:34.060 --> 00:21:47.140 You've got the actual application that needs to be Django or whatever, you know, let's say you've written it in Django and you want your sales site, you know, you've got two 00:21:47.140 --> 00:21:50.420 sites, you've got the actual application and then you have your like sales site. 00:21:51.280 --> 00:21:55.300 If they look really close, then there's no, the transition's really easy. 00:21:55.500 --> 00:22:01.880 So you could have a lot of your blog content or whatever on the static side. 00:22:01.880 --> 00:22:13.080 You're trying to hit lots of random traffic and then, and then flip over to your application and, and the user won't notice that much of a difference because the, it's the same look and stuff. 00:22:13.380 --> 00:22:21.880 So that's, yeah, that's a perfect use case or like the docs for your SAS, you know, like user manual, all that stuff could just all be static, static, static. 00:22:22.160 --> 00:22:22.360 Yeah. 00:22:22.560 --> 00:22:22.940 That's pretty cool. 00:22:22.940 --> 00:22:23.720 I'll give you one more example. 00:22:23.720 --> 00:22:32.920 I talked to, to David Flood from the Harvard sort of humanities enablement team, I guess is a simple way to put it. 00:22:33.180 --> 00:22:44.400 And he and his team go around sort of like on internal consultants or whatever for different researchers who are not really programmers or web people and help them set up websites, 00:22:44.800 --> 00:22:48.220 portals, data management, data exploration, et cetera. 00:22:48.340 --> 00:22:59.580 And one of the areas that he talked to a lot about is like, what happens if you create something like a Django site or anything else as part of a grant, a three-year research project, but then the grant is over. 00:22:59.700 --> 00:23:05.640 You're paying a hundred dollars a month to host the Django site, but now who's going to pay for that, right? 00:23:05.740 --> 00:23:11.460 You're, you don't want to pay for it because you don't have any grant money left, but if you don't, it's going to go away, which is a big hassle. 00:23:11.840 --> 00:23:11.960 Yeah. 00:23:12.060 --> 00:23:15.020 That'd be cool to just be able to convert it to a static. 00:23:15.200 --> 00:23:19.260 It's not going to update anymore and it can be something you can reference and stuff. 00:23:19.700 --> 00:23:19.640 So. 00:23:19.920 --> 00:23:20.360 Yeah, exactly. 00:23:20.480 --> 00:23:28.180 So you've got this project that needed a dynamic site for a long time, but then all of a sudden it no longer is supported or whatever, but you don't want it to vanish from the internet. 00:23:28.560 --> 00:23:32.180 So hit it with one of these, you know, convert to static sites. 00:23:32.240 --> 00:23:33.400 I think it's a really cool idea. 00:23:33.680 --> 00:23:33.940 Yeah. 00:23:34.240 --> 00:23:34.660 It is cool. 00:23:34.880 --> 00:23:35.080 Yeah. 00:23:35.320 --> 00:23:35.680 Indeed. 00:23:35.960 --> 00:23:37.040 You know what else is cool? 00:23:37.440 --> 00:23:38.220 Extras are cool. 00:23:38.600 --> 00:23:38.760 Yeah. 00:23:40.360 --> 00:23:42.460 I'll let you go, go next on the extras. 00:23:42.800 --> 00:23:43.120 Okay. 00:23:43.120 --> 00:23:46.240 I need, I like accidentally unshared. 00:23:46.560 --> 00:23:50.580 So I'll, I'll share it just in, I'll be just a second and here we go. 00:23:51.060 --> 00:23:51.440 All right. 00:23:51.480 --> 00:23:55.800 So this is, is we were talking about, thinking about AI a lot. 00:23:55.900 --> 00:24:01.920 I know a lot of you have kids out there, or maybe you are, going to college or high school or something like that. 00:24:02.000 --> 00:24:04.060 and if you are, thanks for listening. 00:24:04.140 --> 00:24:04.680 That's awesome. 00:24:05.220 --> 00:24:07.480 Anyway, I ran across this article this morning. 00:24:07.620 --> 00:24:13.920 It's called thinking less, trusting more gen AI's impact on students, cognitive habits. 00:24:14.000 --> 00:24:15.740 And actually this isn't an article. 00:24:15.740 --> 00:24:19.660 This is a, an abstract from a study from Oregon state university. 00:24:19.660 --> 00:24:31.480 and like actually the top, there's like a objective and a method and result at the top, but it's rather, it's rather it's, it's not too hard to decipher the results 00:24:31.480 --> 00:24:36.940 looking at this, but it's still in like, you know, it's, it's a, scientific paper speak. 00:24:36.940 --> 00:24:43.740 So, if you pardon the flyby, the best way to read this is to go all the way down to the bottom. 00:24:43.900 --> 00:24:47.340 Oh, not all the way to the, to the references, but there's a conclusion. 00:24:47.720 --> 00:24:50.360 And this is, this is actually interesting. 00:24:50.500 --> 00:24:54.020 I think it's really interesting about students using AI and stuff. 00:24:54.020 --> 00:24:56.600 So I'll just read out the little bit here. 00:24:57.020 --> 00:25:07.300 If routine reliance on gen AI during formative years, and I don't remember what formative years are, but whatever, changes students' willingness to engage in effortful thinking. 00:25:07.440 --> 00:25:19.060 Many may enter professional life without having developed the intellectual habits that earlier generations developed through practice that yet this, this need, this need not be inevitable. 00:25:19.060 --> 00:25:21.440 cognitive, the cognitive debts. 00:25:21.620 --> 00:25:35.080 Anyway, basically, the use of AI and the way AI is designed, like things like chat, GPT and stuff need to be modified so that, learn both learning environments and gen 00:25:35.080 --> 00:25:42.440 AI systems are modified to preserve human agency and support genuine complimentary, complimentarity. 00:25:42.840 --> 00:25:44.600 I didn't know that was a word. 00:25:44.600 --> 00:25:53.060 generally division of cognitive labor between both human and AI contributions should be meaningful, meaningful setup. 00:25:53.300 --> 00:25:55.100 And so that neither is displaced. 00:25:55.200 --> 00:25:58.240 Basically you need to, and I kind of agree. 00:25:58.240 --> 00:26:09.160 I think even when using, like Claude and stuff like that, I think that you can get them to ask you, you can say, Hey, do this task. 00:26:09.160 --> 00:26:11.180 And if you run into issues, ask me about it. 00:26:11.380 --> 00:26:12.580 I think that should be the default. 00:26:12.580 --> 00:26:22.200 I think it should, it should default to, even if, even if you don't want to get interrupted, just like, you know, you want it to have like flow mode and just like get stuff done. 00:26:22.280 --> 00:26:25.800 It should pull out points and say, Hey, these are some of the decisions I made. 00:26:26.080 --> 00:26:29.160 could you review this spot, this pot, this spot? 00:26:29.400 --> 00:26:30.920 and what do you think of that? 00:26:30.920 --> 00:26:32.800 And we could change it, change it if necessary. 00:26:32.800 --> 00:26:38.500 And I think there's minor tweaks that could be done with a lot of these interfaces to keep, people thinking. 00:26:38.500 --> 00:26:41.140 And this is about high school students or about students. 00:26:41.260 --> 00:26:51.960 I don't know what age of students it is, but I think it's relevant for coders also, because it's really easier for coders to slip into trusting, trusting the bots to let it happen. 00:26:51.960 --> 00:26:59.240 And I, we, I am all for getting things done faster, but we need to keep people thinking about the problems and thinking about stuff. 00:26:59.240 --> 00:27:02.580 So anyway, I have a couple of thoughts. 00:27:02.700 --> 00:27:06.500 So I totally agree that this is a very, very, very serious danger. 00:27:06.840 --> 00:27:12.500 I mean, I just think back to myself, if I had a magic box when I was working on a history paper and I could just say, help me with this. 00:27:12.740 --> 00:27:13.500 I really don't remember. 00:27:13.820 --> 00:27:16.400 I don't want to read the 30 page paper. 00:27:16.720 --> 00:27:21.340 And if I, you know, I just, I think it can seriously stunt your, your growth as a student. 00:27:21.340 --> 00:27:29.740 And I honestly think one of the important skills in life is just to learn, to keep going, go, you find a problem, you're like, oh, that's hard. 00:27:29.840 --> 00:27:31.060 Or I can't figure this out. 00:27:31.120 --> 00:27:35.440 Like a huge skill is just learning to keep trying until you do figure it out. 00:27:35.640 --> 00:27:41.040 And if you have a magic box that gives you the answer and you're, especially if you're young, like, I mean, that's really, really rough. 00:27:41.400 --> 00:27:41.500 Yeah. 00:27:41.780 --> 00:27:45.580 You can use these things as really amazing tutors though. 00:27:45.900 --> 00:27:47.640 You can say, I saw you did this. 00:27:48.020 --> 00:27:49.100 I, that's new to me. 00:27:49.120 --> 00:27:49.900 Why did you do this? 00:27:49.920 --> 00:27:51.160 Let's explore this idea. 00:27:51.160 --> 00:27:54.900 But I really doubt many students are using that. 00:27:55.000 --> 00:27:56.380 Like once they get an answer, probably good. 00:27:56.980 --> 00:27:57.220 Yeah. 00:27:57.460 --> 00:28:07.040 Well, and, and like, there's ways to, you can, like I said, also, you, there's ways to use AI tools now to, that work the way they should. 00:28:07.160 --> 00:28:10.800 Like one of like in education, you could say, here's my rubric. 00:28:11.120 --> 00:28:21.580 here's in, it's on this, here's my ideas about what I think I should put, but I, I'm just not sure which part of the paper I should put this information. 00:28:21.580 --> 00:28:25.340 Like the, basically using it to be, I'm stuck. 00:28:25.400 --> 00:28:26.840 Just get me unstuck. 00:28:26.840 --> 00:28:29.120 Don't do everything for me, but just get me unstuck. 00:28:29.360 --> 00:28:32.300 and those are great ways to use this. 00:28:32.300 --> 00:28:33.400 So anyway. 00:28:33.500 --> 00:28:33.780 Indeed. 00:28:34.040 --> 00:28:34.440 Indeed. 00:28:34.680 --> 00:28:36.880 Now, final, final thought before we move off this really quick. 00:28:37.140 --> 00:28:39.920 There's a lot of EM dashes in this. 00:28:40.660 --> 00:28:46.020 I'm pretty sure that this entire abstract was written with AI or at least rewritten with AI. 00:28:46.160 --> 00:28:48.440 Like just on that page, there's three or four EM dashes. 00:28:48.440 --> 00:28:54.340 But I'm not against EM dash, but that is such a new thing that AI seemed to have brought into the mix. 00:28:54.780 --> 00:28:54.920 Right. 00:28:55.040 --> 00:29:01.860 But, but there's what the, what the grammarly has been used commonly forever for a long time then. 00:29:01.960 --> 00:29:05.560 And I don't think that like that grammarly counts as an evil thing. 00:29:05.760 --> 00:29:06.580 yeah, I don't either. 00:29:06.880 --> 00:29:15.080 So, but, also there's way more EM dashes in, in it's like research papers than there are normal papers. 00:29:15.080 --> 00:29:15.620 That is true. 00:29:15.620 --> 00:29:18.240 And honestly, that's probably how they got into AI, but still. 00:29:18.440 --> 00:29:18.800 Probably. 00:29:19.160 --> 00:29:23.500 I'm still thinking, I'm still thinking perhaps that I got a couple of things. 00:29:23.680 --> 00:29:26.140 some of these are a little, this is a little bit older. 00:29:26.220 --> 00:29:26.920 It's 14 days ago. 00:29:26.920 --> 00:29:30.540 Cause it came out right after we recorded our last episode and then we skipped a week. 00:29:30.600 --> 00:29:33.760 So it's not that new, but it's already, this bothers me. 00:29:33.760 --> 00:29:36.820 Just the first four, four words of this announcement. 00:29:36.920 --> 00:29:37.260 Bother me. 00:29:37.500 --> 00:29:39.020 AI cloud company Vercel. 00:29:39.180 --> 00:29:43.660 Wait, no, this is a cloud company that hosts infrastructure, but they probably wanted to raise more money. 00:29:43.660 --> 00:29:46.780 So now they're an AI cloud company, not just a cloud company, but whatever. 00:29:47.080 --> 00:29:48.240 Let's just start from after that. 00:29:48.440 --> 00:29:48.920 Vercel. 00:29:49.920 --> 00:29:57.400 Vercel is breached after an employee grants an AI tool unrestricted to Google works, unrestricted access to Google workspace. 00:29:57.740 --> 00:29:59.600 Hacker seeks $2 million for stolen data. 00:29:59.600 --> 00:30:12.760 So some, I got to go through a little bit more, but, basically someone was using agentic AI and gave it access to the production Google workspace, which stored a bunch of data and probably wikis and stuff like that. 00:30:12.760 --> 00:30:18.780 And then somehow they installed something bad that then used the AI, which used that access. 00:30:18.780 --> 00:30:21.500 I don't know exactly the details, but why do I bring it up? 00:30:21.560 --> 00:30:26.660 If you're using Vercel, you might want to rotate your keys, change your password, stuff like that. 00:30:26.880 --> 00:30:27.700 I don't think it's a mispronation. 00:30:27.700 --> 00:30:29.140 Vercel is seeking the money though. 00:30:29.420 --> 00:30:29.680 Yes. 00:30:29.680 --> 00:30:32.740 Cause they stole the data from them and they'll say, I'm going to release it to the world. 00:30:32.740 --> 00:30:35.240 If you don't pay me, it's like a ransomware sort of thing. 00:30:35.240 --> 00:30:35.600 Okay. 00:30:35.620 --> 00:30:35.920 Yeah. 00:30:35.920 --> 00:30:36.200 Yeah. 00:30:36.200 --> 00:30:36.360 Yeah. 00:30:36.460 --> 00:30:36.760 Okay. 00:30:36.820 --> 00:30:37.240 Got it. 00:30:37.340 --> 00:30:37.940 No, the M dash. 00:30:38.040 --> 00:30:38.240 Okay. 00:30:40.640 --> 00:30:43.880 I mean, this is the reason there's just another warning for everyone. 00:30:43.880 --> 00:30:53.360 Like you got to be really careful with your dev machines, which are often for a lot of us also our personal computers, you know, like the whole, what was the password manager? 00:30:53.580 --> 00:31:04.000 Not one password, not pit warden last pass last pass got hacked, not because of an AI thing, but because one of the devs had a home network, which had, a Plex server. 00:31:04.000 --> 00:31:08.860 So they could stream their pirated stuff to their, just to watch. 00:31:08.860 --> 00:31:09.140 Right. 00:31:09.320 --> 00:31:11.880 But then they want to do that while they were out, while they're traveling. 00:31:11.880 --> 00:31:13.020 So they put it on the internet. 00:31:13.020 --> 00:31:20.540 It got hacked and then it laterally moved through their home network to get to their computer, which then got all the API keys and on and on it goes. 00:31:20.660 --> 00:31:21.960 So anyway, just another little warning. 00:31:22.340 --> 00:31:24.080 And if you use Vercel, maybe a double warning. 00:31:24.460 --> 00:31:24.920 All right. 00:31:25.100 --> 00:31:25.860 That was kind of sad. 00:31:25.960 --> 00:31:26.740 I'll give you a good one, Brian. 00:31:26.840 --> 00:31:27.300 Check this out. 00:31:27.480 --> 00:31:28.660 This is very relevant. 00:31:28.660 --> 00:31:34.820 So I want to, I went and I was just thinking like, I really want a better experience for people over at talk Python. 00:31:34.820 --> 00:31:38.240 And I'm doing a crazy amount of stuff with transcripts. 00:31:38.280 --> 00:31:40.920 Like I'll talk about it next week, probably. 00:31:41.480 --> 00:31:47.620 But for weeks I've been working on making transcripts better, more with the courses than with the podcast. 00:31:47.620 --> 00:31:49.380 But still it applies to both. 00:31:49.620 --> 00:31:53.760 And like, it would be really nice if there was a better playback experience on the web. 00:31:53.780 --> 00:31:57.740 Like imagine this, you come to the website, Python Bytes or talk Python, whatever. 00:31:57.740 --> 00:31:59.140 We've got really nice search. 00:31:59.220 --> 00:32:03.220 And I've talked about the work I've done on the Python Bytes search, which I think is really cool. 00:32:03.340 --> 00:32:08.800 Some of the ways you can search and jump around, but maybe you find something interesting from 200 episodes ago. 00:32:09.040 --> 00:32:16.640 Maybe you could get that on your podcast player, but you've got to like scroll back and like, how, you know, does your player cut off episodes older? 00:32:16.880 --> 00:32:18.000 And like, it's just not great. 00:32:18.040 --> 00:32:20.460 You might just want to watch it on the, listen to it on the web, right? 00:32:20.700 --> 00:32:20.860 Yeah. 00:32:21.080 --> 00:32:22.780 So you search for it, you click, you start going. 00:32:22.780 --> 00:32:31.240 And maybe you could have a better experience than just the default playback, you know, audio tag that for some reason browsers think doesn't deserve much love or interest. 00:32:31.380 --> 00:32:32.400 They're like, they're crappy. 00:32:32.500 --> 00:32:35.840 They look like they're from, you know, real player in 95, but whatever. 00:32:35.840 --> 00:32:42.780 So I went and I created a really nice playback experience where, let me just jump over. 00:32:43.260 --> 00:32:45.440 I'll link to the blog post here, but come in here. 00:32:45.600 --> 00:32:48.840 It's got nice little access to all the pieces, but it's got a player. 00:32:48.920 --> 00:32:52.740 If you click it, it expands out with like little seek stuff around. 00:32:52.880 --> 00:32:58.020 It shows the transcripts as it's playing, like right alongside the player. 00:32:58.140 --> 00:32:59.500 You can like change the speed. 00:32:59.500 --> 00:33:01.720 It's got hotkeys, all sorts. 00:33:01.920 --> 00:33:02.400 And that's slick. 00:33:02.760 --> 00:33:03.060 Oh yeah. 00:33:03.160 --> 00:33:06.700 And thanks for doing the speed thing too, because I'd never listened at 1x. 00:33:07.060 --> 00:33:07.360 Yeah. 00:33:07.480 --> 00:33:07.720 Yeah. 00:33:07.900 --> 00:33:10.040 So Python bytes go over here. 00:33:10.200 --> 00:33:11.680 I've got the same deal. 00:33:11.740 --> 00:33:14.640 I just didn't write a blog post about it because I did talk Python first, but look at that. 00:33:14.800 --> 00:33:16.240 Oh, very cool. 00:33:16.580 --> 00:33:17.080 Isn't that nice? 00:33:17.340 --> 00:33:22.280 So people out there listening, if you're like, I'm just kind of exploring our content. 00:33:22.360 --> 00:33:23.820 I mean, we've got a lot of years. 00:33:24.500 --> 00:33:25.860 How many years, what was our first episode? 00:33:25.920 --> 00:33:26.860 Let me just look really quick, Brian. 00:33:26.900 --> 00:33:27.940 What was our very first episode? 00:33:27.940 --> 00:33:30.080 It was 2016. 00:33:30.260 --> 00:33:32.820 That's 10 years of content. 00:33:33.080 --> 00:33:33.500 Yeah. 00:33:33.640 --> 00:33:36.240 We've got a great search engine and transcripts on top of that. 00:33:36.360 --> 00:33:38.540 And so if you're exploring that, you know, just use the player. 00:33:38.640 --> 00:33:39.540 I think it's really, really cool. 00:33:39.740 --> 00:33:41.420 We've been doing this for 10 years almost. 00:33:41.640 --> 00:33:42.260 I know. 00:33:42.700 --> 00:33:43.460 Damn, you're old. 00:33:43.960 --> 00:33:49.520 Dude, weird how like some people can get old and others in the same time frame just don't, right, Brian? 00:33:49.780 --> 00:33:50.220 Yeah. 00:33:50.620 --> 00:33:50.760 Yeah. 00:33:50.900 --> 00:33:51.460 Well, I guess. 00:33:52.240 --> 00:33:59.120 So you'll, this is funny because if you're watching this, you'll know it, that I'm the one with gray hair, not Michael. 00:33:59.460 --> 00:34:00.480 But you got more hair. 00:34:00.600 --> 00:34:01.500 So I think it evens out. 00:34:01.740 --> 00:34:03.940 So thank you. 00:34:03.980 --> 00:34:06.340 Pat says last pass reminds me of as I was stumbled on. 00:34:06.460 --> 00:34:06.680 Okay. 00:34:06.920 --> 00:34:08.060 A couple more things real quick. 00:34:08.600 --> 00:34:10.900 A combination of stuff in this last item. 00:34:10.900 --> 00:34:18.200 I both want to give a little bit, a little bit of sympathy to GitHub and then also kind of bash on him a little bit. 00:34:18.540 --> 00:34:20.140 So GitHub has been having. 00:34:20.240 --> 00:34:21.400 As we do, you know. 00:34:21.800 --> 00:34:22.920 As I think it's fair. 00:34:23.020 --> 00:34:24.120 Like, isn't that balanced? 00:34:24.500 --> 00:34:24.620 Yeah. 00:34:24.620 --> 00:34:27.500 I think you just got to say if there's two sides, you got to present them both equally. 00:34:27.640 --> 00:34:29.980 But there's, there's a little bit of nuance here. 00:34:30.060 --> 00:34:37.320 So I don't know if you've noticed, but GitHub has been a little janky lately and not in just the GitHub actions, which have been taking a lot of heat, but just uptime at all. 00:34:37.320 --> 00:34:42.780 Like I couldn't, I couldn't search the other day a little while before then, like issues were not accessible. 00:34:43.120 --> 00:34:46.780 I didn't experience this, but people had their pull requests just disappear. 00:34:47.000 --> 00:34:51.340 How frustrating would that be to do some work, submit a pull request and it just goes into the ether. 00:34:51.460 --> 00:34:52.580 You're like, are you serious? 00:34:52.780 --> 00:34:54.760 So here's the supportive angle. 00:34:55.280 --> 00:35:00.940 They, they did this, this post over on the GitHub blog called an update on GitHub availability. 00:35:01.700 --> 00:35:04.740 They have a graph that says things are hard for us. 00:35:04.820 --> 00:35:05.640 We're trying. 00:35:05.940 --> 00:35:07.020 They're doing a couple of things. 00:35:07.020 --> 00:35:11.760 I believe they're reworking GitHub from being a monolith into being a bunch of microservices. 00:35:11.760 --> 00:35:21.220 And I usually hate on microservices, but something like GitHub with all these different moving parts and different levels of traffic and load on them seems like you might want to, well, 00:35:21.220 --> 00:35:29.920 maybe we should just scale up the issues a bit, like a ton, but not so much just listing the repos or whatever. 00:35:30.240 --> 00:35:35.780 So apparently they're going through, they're like, oh, we need something like 10 X more capacity than we had a couple of years ago. 00:35:36.040 --> 00:35:38.940 And during that time, they're like, actually, no, maybe 30 X more. 00:35:39.420 --> 00:35:40.100 So it's a lot. 00:35:40.520 --> 00:35:47.460 And if you look at the graphs, if you look at the graphs, they are nothing but like nearly vertical in the amount of traffic and new repos and commits. 00:35:47.780 --> 00:35:55.740 And one of the main drivers of this is people, especially a lot of new people to programming who were not programmers are like, hey, Claude, create me a project. 00:35:55.780 --> 00:35:57.040 Where do I save my project? 00:35:57.220 --> 00:35:58.080 Oh, you put it on GitHub. 00:35:58.340 --> 00:35:58.920 You know what I mean? 00:35:58.920 --> 00:36:04.020 And so just the amount of traffic amplified by AI to GitHub is over the top. 00:36:04.280 --> 00:36:04.420 All right. 00:36:04.620 --> 00:36:14.460 So that explains- You know, maybe if we could charge like 50, if they charge like 50 cents a month, just something small, just, but anything like a dime a month, a lot of that would go away. 00:36:14.700 --> 00:36:15.540 That's really interesting. 00:36:15.700 --> 00:36:16.960 I would totally pay a dime a month. 00:36:17.060 --> 00:36:19.340 I think I pay $30 a month to get up. 00:36:19.540 --> 00:36:22.520 I wouldn't mind paying 30 and 10, 30 point, $30, 10 cents. 00:36:22.760 --> 00:36:23.500 Wouldn't make a difference. 00:36:24.360 --> 00:36:27.120 Anyway, no, I think that's a very interesting point, actually. 00:36:27.120 --> 00:36:27.560 Okay. 00:36:27.560 --> 00:36:28.520 So there's that. 00:36:28.860 --> 00:36:30.280 This is GitHub's response. 00:36:30.620 --> 00:36:33.060 Did you know that there's a missing GitHub status page? 00:36:33.460 --> 00:36:34.280 Yes, there is. 00:36:34.560 --> 00:36:37.780 If you go back to the previous one, somewhere did they talk about their uptime? 00:36:38.160 --> 00:36:40.360 How could it exist if it's missing? 00:36:40.740 --> 00:36:42.400 So is there a status? 00:36:42.840 --> 00:36:43.020 Yeah. 00:36:43.060 --> 00:36:47.300 So if you go to GitHub status page, it says, oh, Git operations are up, up, up. 00:36:47.480 --> 00:36:48.580 Web hooks are up, up, up. 00:36:48.620 --> 00:36:54.380 And it talks about like, it doesn't aggregate it, but it's like 99.5 sort of uptime. 00:36:54.380 --> 00:36:58.500 But then there's another page, which ironically is hosted on GitHub.io. 00:36:58.660 --> 00:36:59.200 I love it. 00:36:59.480 --> 00:37:00.820 It's like its own thing. 00:37:01.260 --> 00:37:04.280 And it basically does other types of checks. 00:37:04.440 --> 00:37:07.300 And it's like, you know, 84% uptime. 00:37:07.460 --> 00:37:09.440 That is actually pretty abysmal. 00:37:09.800 --> 00:37:10.040 Yeah. 00:37:10.300 --> 00:37:11.340 And it just has graphs. 00:37:11.560 --> 00:37:13.840 And it actually breaks out pull requests and all these things. 00:37:14.140 --> 00:37:16.200 And here you look and you see kind of similar numbers. 00:37:16.320 --> 00:37:17.700 It's like the Git operations are this. 00:37:17.940 --> 00:37:18.720 Web hooks are this. 00:37:18.720 --> 00:37:23.180 But I think it says, if any of these are down, then you don't call it up. 00:37:23.580 --> 00:37:23.960 You know what I mean? 00:37:24.280 --> 00:37:26.940 Whereas the other one goes, if some of it is up, it's up. 00:37:27.580 --> 00:37:28.460 Not all of it. 00:37:28.520 --> 00:37:30.380 I mean, you don't like issues right now, do you? 00:37:30.640 --> 00:37:33.340 I don't know exactly the breakdown there. 00:37:33.400 --> 00:37:36.400 But I just thought this missing GitHub status page was really interesting. 00:37:36.760 --> 00:37:37.020 Yeah. 00:37:37.200 --> 00:37:38.120 They should do better. 00:37:38.420 --> 00:37:41.380 I mean, I know that a lot of us are using it for free. 00:37:42.240 --> 00:37:46.800 But also, we've paid for it with all of our intellectual property. 00:37:46.800 --> 00:37:48.640 Yeah, exactly. 00:37:49.080 --> 00:37:49.800 I mean, that's... 00:37:50.420 --> 00:37:51.460 And attention, right? 00:37:51.520 --> 00:37:58.120 The reason that there's not a mass exodus to something new or shinier is because this is where most of the people are. 00:37:58.260 --> 00:38:01.540 I know there is something of an exodus, but not a mass exodus, right? 00:38:01.780 --> 00:38:02.080 Yeah. 00:38:02.460 --> 00:38:04.300 You get benefits from being on GitHub. 00:38:04.580 --> 00:38:04.880 Okay. 00:38:05.420 --> 00:38:11.880 I think all of these conversations about these AI things and stuff and others perfectly lead into this joke. 00:38:11.980 --> 00:38:14.800 And this is probably one of my favorite jokes. 00:38:14.800 --> 00:38:15.080 Okay. 00:38:15.080 --> 00:38:17.400 At least of the year. 00:38:17.520 --> 00:38:18.220 Have you looked at it yet? 00:38:18.480 --> 00:38:18.720 No. 00:38:19.060 --> 00:38:19.300 Okay. 00:38:19.420 --> 00:38:19.780 Perfect. 00:38:19.980 --> 00:38:20.160 Okay. 00:38:20.540 --> 00:38:21.560 So I found this on X. 00:38:21.980 --> 00:38:22.900 And I'm programming humor. 00:38:23.040 --> 00:38:24.260 I guess it could also be on Reddit. 00:38:24.400 --> 00:38:26.500 Like they broadcast that in multiple places. 00:38:26.920 --> 00:38:30.640 So this is like two different perspectives from society. 00:38:31.000 --> 00:38:33.160 And it's talking about Copilot, but it could be ChatGPT. 00:38:33.240 --> 00:38:34.040 It could be Claude, whatever. 00:38:34.380 --> 00:38:35.340 Friends outside of tech. 00:38:35.580 --> 00:38:36.980 Lol, Copilot is dumb. 00:38:37.260 --> 00:38:37.820 Friends in tech. 00:38:38.060 --> 00:38:39.440 I just bought iodine tablets. 00:38:39.640 --> 00:38:41.220 I have made an offer on land upstate. 00:38:41.220 --> 00:38:47.020 My supplies of antibiotics and potable water are sufficient, but I need to set up hydroponics to make it through the first few years. 00:38:49.180 --> 00:38:51.740 Did that just perfectly sum it up? 00:38:53.100 --> 00:38:53.260 Yeah. 00:38:53.340 --> 00:38:54.980 It's like, this is a dumb thing. 00:38:55.040 --> 00:38:56.800 Look, it made this thing longer. 00:38:56.800 --> 00:38:58.800 We're like, I'm almost done stockpiling. 00:38:58.900 --> 00:39:00.020 We're ready to build the bunker. 00:39:00.820 --> 00:39:01.840 I know what is coming. 00:39:01.980 --> 00:39:02.600 I can see it. 00:39:02.660 --> 00:39:03.260 It's so bad. 00:39:04.080 --> 00:39:04.300 Yeah. 00:39:04.660 --> 00:39:08.880 I'm, I've just like, I've just, I know that I'm dependent on society. 00:39:09.000 --> 00:39:15.640 If society collapses, just, I hope I'm one of the first to get hit because I don't want to live through an apocalypse. 00:39:16.180 --> 00:39:17.220 I really don't either. 00:39:17.520 --> 00:39:18.860 I'm not a survivalist. 00:39:18.860 --> 00:39:25.400 So, well, stock up on your, iodine tablets and your potable water. 00:39:25.560 --> 00:39:26.300 Oh my God. 00:39:26.400 --> 00:39:27.060 It's so bad. 00:39:29.040 --> 00:39:39.060 Well, so I've got, I've got a friend that like, they've got like a group of friends that have a place like in Southern Oregon that they, they know they can hole up in and stuff. 00:39:39.520 --> 00:39:40.780 If anything happens. 00:39:40.780 --> 00:39:41.960 And I said, oh, okay. 00:39:41.960 --> 00:39:45.660 So, so do you make sure that you always have a full tank of gas? 00:39:45.680 --> 00:39:47.580 So if anything happens, you can make it there. 00:39:47.580 --> 00:39:48.320 Oh no. 00:39:48.680 --> 00:39:51.120 Well, how are you going to get there? 00:39:51.340 --> 00:39:53.000 Are you going to walk to Southern Oregon? 00:39:53.440 --> 00:39:53.660 Anyway. 00:39:55.560 --> 00:39:56.000 So. 00:39:56.340 --> 00:39:56.440 Yeah. 00:39:56.540 --> 00:39:57.040 Good point. 00:39:57.240 --> 00:40:01.000 There's a lot of moving pieces and the one you didn't see coming might be the one that takes you. 00:40:01.000 --> 00:40:05.080 I hope he doesn't listen to this, this podcast, but I hope everybody else does. 00:40:05.320 --> 00:40:05.960 Yes, exactly. 00:40:06.360 --> 00:40:10.060 Well, everyone, hope your supplies are ready and thank you for listening. 00:40:10.520 --> 00:40:11.060 Thank you. 00:40:11.360 --> 00:40:11.760 Bye. 00:40:11.820 --> 00:40:12.340 Thanks everybody. 00:40:12.560 --> 00:40:12.780 Bye.